Spinozazian roots of Gestalt isomorphism

Riccardo Luccio
Belgrado, Spring 2009

Introduction
The Gestalt concept of isomorphism has been subject to many misinterpretations. They must be attributed at least in part to some obscurities in classic Gestaltist writings. The first refers to the very name isomorphism. The second refers to the domains that are supposed isomorphic The third ambiguity, inside the consciousness, contrasts the direct experience to the representations. The fourth contrasts structural versus functional isomorphism.

Introduction
The fifth is ontological: why some authors can interpret isomorphism as a dualistic concept? And if it is monistic, it is a neutral monism? Or a materialistic monism? Or an idealistic doctrine? In conclusion, I try to solve all this ambiguities going back to Spinoza, possibly the main source of inspiration, at least of Wertheimer.

A short historical sketch

The origin of isomorphism is in Wertheimers account of the phi phenomenon, with his neurophysiologic hypothesis of the Querfunktionen (cross functions) and of the physiological Kurzschlu (short-circuit) (1912). As any idea in the history of science, the concept of isomorphism too had noteworthy antecedents (Th. Lipps. Lotze, Grassmann).

Hering
Hering is correctly considered one of the most direct precursors of Gestaltpsychologie. At the basis of the idea of Hering (1878) were the concepts of assimilation and dissimilation. Assimilation and dissimilation are well demonstrated in the case of visual sensation. But it would be curious if only the dissimilative side of the process should be influential in the perceptual process. And more curious when this process would be exclusive of vision.

E. E. Mller
The importance as precursor of isomorphism of Georg Elias Mlleris linked to its famous 1896 paper on the five psychophysical axioms (Mller, 1896). For us important re the first three. (1) The ground of every state of consciousness is some material (psychophysical) process. (2) To every equality, similarity, or difference of a sensation corresponds respectively an equality, similarity, or difference of the underlying psychophysical process, and vice versa [umgekehrt].

E. E. Mller
(3) If the changes of a sensation have the same direction, or if the differences between a series of sensations have the same direction, the like will be the case in regard to the corresponding psychophysical process; if a sensation is variable in n directions, so also is its psychophysical process. However, as Vicario (2001, p. 88 f.) points out, the umgekehrt of the second axiom is unnenecessary, unprooved and immotivated.

The phi phenomenon

The idea of Querfunktionen and Kurzschlu went to Wertheimer from observations of several investigators before him: Exner (1875), Marbe (1898), Drr (1900), Wundt (2002), Schumann (1907). In particular, two were the observations that particularly impressed Wertheimer.

The phi phenomenon

The first was Exners aploscopic (with independent stimulation of the two eyes) observation, which confirmed that one had to invoke something beyond the retina to explain the perception of apparent movement. The second derived from a neuropsychological case: Otto Pltzl, an important neurologist from Vienna, had observed a patient with a lesion of both occipital lobes; presenting to her one moving light, she saw a sequence of separate lights.

The phi phenomenon

According to Max Wertheimer (ibidem, p. 247), the present (at the time) physiological research was indeed sufficient to assume as likely that the excitation of a central point a elicits a physiological effect in a definite area around it. When are two the points a and b that are excited, a similar effect in both points should result.

The phi phenomenon

When the point a is excited, and thereafter the point b, within some specifically short time interval, then a sort of physiological short-circuit from a to b should occur. There is a specific passage of the excitation in the space between the two points. If for instance the extent of the disturb in the area around a has reached the maximum of the temporal curve of its process, and the disturb in the area around b takes place now, then the excitation flows (a specific physiological event), and its direction is determined by the fact that the excitation around a occurred first.

The phi phenomenon

It is incorrect to say, as somebody did, that for Wertheimer this approach was axiomatic (Carini, 1970). For Wertheimer this idea was only a sketch of a physiological hypothesis (Skizzierung einer physiologiker Hypothese), virtually subject to experimental testing, as all the scientific hypotheses, and absolutely not an axiom. As we will see later, instead Carini could be in some sense correct in the case of Khler.

The physical Gestalten

Wolfgang Khler fully developed Wertheimer's intuition. His book Die physischen Gestalten in Ruhe und im stationaren Zustand was written when Khler was in Africa, in Tenerife, Canary Islands, during the First World War. When Khler was in Tenerife, he read Electricity and Magnetism by Clerk Maxwell, and he was great relieved to find so fundamentally similar an approach between great physicists and Gestalt psychologists. One must add that in Khlers book was evident the influence of another eminent physicist, Ernst Mach.

The physical Gestalten

Eventually, in 1920 Khler was able to go back in Germany, and his book on physical Gestalts appeared. Khler (1969, p. 77) recalls that when the book was published in 1920, both Wertheimer and Koffka greatly enjoyed its content: It showed that the alleged mysteries of Gestalt psychology agreed with perfect clear procedurs and facts in natural science. In a sense, Gestalt psychology has since become a kind of application of field physics to essential parts of psychology and of brain physiology.

The physical Gestalten

von Ehrenfels (1890) had defined suprasummativity and transposition as key concepts for Gestaltqualitten. The point of departure of Khler consists in indivuating the same properties in an electric field, that is in the distribution of electrical charges around a conductor. The second step is to hypotise that in the brain there are chemicophysical fields having the same properties. The final step is individuating the same systema properties [Systemeigenschaften] in both the domains, the experience (the phenomenal field) and the brain.

The physical Gestalten

Four properties are quite similar in phenomenal and in brain fields: i) ii) The total processes appear in both fields as units with dynamic properties; in both the unity is compatible with a structured articulation [Gliederung] of the component parts;

iii) in both one can individuate gradients due to the distance from one region to another that consent to consider the regions as indipendent from the ones that are far away; iv) in both we can individuate limited regions (Gestalten, in the phenomenal field) on a ground.

The postulate of the isomorphism postulate

Khler in several instances, but in particular in The Place of Values in a World of Facts, says that the thesis of isomorphism [...] is not yet a theory, but remains a postulate. [...] Isomorphism is a postulate. It becomes a theory by virtue not of one hypothesis, but a whole set of definite assumptions.

Now, in logic as in mathematics axiom and postulate are almost synonims, to design a proposition not proved, but self-evident, which truth is taken for granted.

The postulate of the isomorphism postulate

In other words, what I can understand in what Khler is saying here is that isomorphism is a logical necessity, or something of similar. On the basis of this necessity, we need to formulate some assumptions, that will have the status of scientific hypotheses to test experimentally, and so the postulate will transform itself in a theory. Anyway, if this is Khlers real position it is difficult to contradict Vicario (2001), when he states that the isomorphism is maintained not on the basis of a scientific need, but for metaphysical urges.

Why isomorphism? isomorphism

The term isomorphism was used first by Khler only in 1929. Khler himself used in his writings the term with great parsimony. In my opinion, the choice of this term was really unfortunate. Indeed, it had a great success, but this success was also the source of plenty of misunderstandings, abuses, misuses in improper areas, and so on. In abstract algebra we say that between two domains there is an isomorphism if there exists bijective morphism, that is a preserving structure mapping.

Why isomorphism? isomorphism

In mathematics the concept of isomorphism has two components (Madden, 1957). Given the two domains, for whom must exist a definite set of axiom ruling the relations of the elements inside them, to saying that there is an isomorphism there must exist a correspondence one-toone between the elements of the two domains, such that to any element of the first corresponds one and only one element of the second, and vice versa. The system of axioms specifies in each domain their internal structure.

Why isomorphism? isomorphism

In all that Gestalt authors say about the relations between the domains, the first aspect of the isomorphism is not respected, so the use of the term is under this aspect inappropriate. Wertheimer, (in letter written in the thirties to Boring) says Now, we have never asserted isomorphism in the sense, defined by you, of a one-one correlation. Luchins and Luchins (1999, p. 208) state:In Gestalt psychology, the one-to-one correspondence between elements is not required; similarity of structures is required.

Why isomorphism? isomorphism

Khler (1920, p. 193) hypothesises a sort of material similarity (sachlicher Ahnlichkeit) between psychophysical process and phenomenal field, as far as their Gestalt properties are concerned, that refers to what Khler calls structural. In other words, both in the phenomenal field and in the brain processes could exist other, but the isomorphism applies onlyas far as Gestalt properties are concerned.

Why isomorphism? isomorphism

However, in the last decades, thanks to the popularity gained by the term isomorphism", we have seen in the cognitive field an almost unlimited proliferation of proposals of isomorphism of all sorts. The most popular is Shepards second-order isomorphism, but we have also also Vanderverts "isomorphism in NP", Uttals dimensional isomorphism" (or nonisomorphism"), Pessoa and colls analytic isomorphism, Palmers natural isomorphism or isomorphic constraint, Wrights structural isomorphism; and the list could be longer.

The domains
We said at the beginning that usually, out and inside the Gestalt psychology, the domains that are considered isomorphic can be three: the domain of the real world, the psychological domain of the phenomenal world, and the physiological domain. The problem that arises is that these domains are less clearly defined than one can imagine, and this not in the interpretations of the scholars external at the field, but in the writings of the Gestalt psychologists themselves.

The domains
As for the external scholars, for instance Hatfield (2003) attributes a threefold model to Khler(1929): He proposed, as an empirical thesis, that during perception the shape of a spatial structure in the world causes brain events exhibiting similar shape (isomorphism means same shape), which yield phenomenal presentation in experience of that shape (as seen from a point of view).

The house that Jack built

But, as a matter of fact, this is not what Khler says, but what Shepard (1981) (that Hatfield quotes immediately after) says about what Khler says. And, in turn, this is what Skinner (1963, that Shepard, p. 284, quotes immediately after) says that Khler says. This is what happens in the house of the isomorphism that Jack built out of Gestalts quarter. As correctly Henle (1984) states, this implies the identification of perceptual and physical objects (nave realism), that makes the discussion irrelevant to the Gestalt concept of isomorphism.

Wertheimer Wertheimers v. Khlers isomorphism Khlers

According to Newman (1989) Wertheimer was a deeply committed monist, that always believed in the physiological hypothesis. However, Luchins and Luchins (1999) suggest that Wertheimer was more oriented than Khler to experimental phenomenology and less interested in physiological hypotheses. Such difference might help to account for differences in their conceptions of isomorphism.

Wertheimer Wertheimers v. Khlers isomorphism Khlers

According to Luchins and Luchins, during the 1937-38 seminars Wertheimer defined the isomorphism as the thesis that the Gestalt quality of psychological events is similar to the quality of the physical world. Wertheimer admitted that his own formulation differed from Khlers one, and gone further saying that his own isomorphism does not always hold. In my opinion here we have at most a very weak meaning of the term.

Arnheim Arnheims isomorphism

According to Arnheim (1987, p. 210) the concept of isomorphism is not limited to the relation between perceptual phenomena and the corresponding neurological processes. It refers to any structural analogy between patterns located in different media, and in this broader sense it is of central relevance to gestalt theory itself. For instance, when one tries to understand how the melodic and harmonic patterns of music or the motions of a dancer convey their expressive meaning with such striking immediacy, the isomorphic relation between the shapes and colours perceived and the meanings transmitted by them is the only base from which a concrete analysis of practical examples can be undertaken.

... and Metzgers Gestaltverwandschaft Metzger

Some years later, Arnheim (1949) developped further his concept of isomorphism, envisaging different isomorphic levels when more than one observer is implied. A similar interpretation of the isomorphism appears also in other authors, e. g. Hartmenn (1935) and S. E. Asch (1958). However, Metzger (1941) says that in such cases of whom Arnheim speaks (that is not quoted by Metzger, of course; and conversely), one should speak of Gestalt affinity [Gestaltverwandschaft], Here one can hypothesize that there is a causal chain, but one must exclude a structural identity a true isomorphism.

The problem of the emergence emergence

According to Khler the real problem is to interpret phenomenal facts in isomorphic terms, avoiding the drawback of the parallelism, utterly rejects. The laws that rule the natural events are invariant, and leave no room to evolutionary changes. This postulate of invariance is in sharp contrast with every possible emergentist stance.

The problem of the emergence emergence

As Khler (1960, p. 20-21) says, Nobody who has studied physics would predict that, when physical processes occur in brains, they will affect events in an entirely different realm, the mental world, and that in turn, they will be affected by such mental events. [...] From the point of view of science, causal relations of this kind would have to be regarded as instances of an incomprehnsible emergence. [...] Parallelists ['...] view also implies that what happens in brains when we perceive, feel, think, and so forth, is a novum from the standpoint of natural science [...] this once more means emergence.

Consciousness v. phenomenal world

Another point is relative to the second domain: consciousness or phenomenal world? In my opinion, if we consider the years in which the concept was developed, it is clear the consciousness refers to another field of theorising. Bissett Pratt could write (1936, p. 155), the identity of physiological processes with subjective states, observed by introspection, is sometimes maintained by [Gestalt psychology] on the basis of their isomorphism. Of course, this is not what Gestalt psychologists maintain.

Sense organs v. CNS

Includes the physiological domain also the peripheral level of the sense organs, or is it limited to the central brain functions? In many occasions Gestalt authors say overtly that the selforganisation is possible only at the central level, and not at the level of the proximal stimuli. For example, Khler (1966, p.76) say that local retinal stimulations are mutually independent facts, just as the elements of all physical surfaces before the organisms reflect light independently, then these stimulations must be regarded as a mere mosaic.

Sense organs v. CNS

So, it is a little surprising that somebody advocates a second meaning to isomorphism, which focuses on the structural resemblance between the proximal stimulus (retinal image) and its representation on the cortical surface (Ehrenstein, Spillmann, and Sarris, 2003, p. 446). Of course, this kind of isomorphism is congruent with some recent developments in neuroscience, but this has little to do with Gestalt psychology.

Is Gestalt psychology representationalist?

The phenomenal field is what, of whom the subject is directly conscious the content of consciousness. Could this mean that this is the domain of the representations? This is what some author believes, and in a very clear way is what Lehar (2003, a and b) and Scheerer (1994) maintain.

Is Gestalt psychology representationalist?

The term representation has a clear story with a long philosophical past, and an intense contemporary life. In English representation means solely something staying for something other (see Hubbard, 2006) gives of it: A representationalist view is a view, in which in the mind of the subject there are representations that stay for the events of the internal and/or external world.

Is Gestalt psychology representationalist?

Of course, as in Haselager, De Groot, and van Rappard, H. (2003), one can build an isomorphistic representational theory, but it is not a Gestaltist theory. Or one can propose an anti-representationalist theory that can be isomorphist or not, as in the cases of the proposals of smart mechanisms: Polar Planimeter for Runeson (1978), Watt Governor for van Gelder (1995); but again, Gestalt theory is another matter.

Is Gestalt psychology representationalist?

We must also consider that there is a tradition in European psychology of this concept. In particular, there is a line that goes from Brentano, to Graz school, to Berlin school. Now, the concept of representation (and of presentation) is very important in Brentanos system, linked to the Thomistic concept of intentional in-existence (Brentano, 1874). But Gestalt theorists rejected clearly this part of Brentanos theory as they rejected Meinongs production theory. Another important remark. Often, in European tradition, we use the term henomenal field to indicate the econd

Is Gestalt psychology representationalist?

Khler, who speaks of directly accessible world (1966, p. 73), says that another name of this part is the phenomenal world. I have not used this name because it almost sounds as though it referred to mere appearances of other, more substantial, and therefore more important, facts. I think that this is the eventual tombstone of any attempt to qualify Gestalttheorie as a representational theory.

Is Gestalt psychology representationalist?

We can ask to ourselves why, if Gestalttheorie is a representational theory, the authors almost never use the term representation, or its German numerous synonyms. As an exception: a book by Koffka (1912) on visuelle Vorstellungen, theoretically far from Gestalttheorie. Note that the term could be used without any theoretical commitment to a representational theory. But always the Gestalt psychologists preferred other terms. For instance, in the paper on thinking of primitive peoples, referring to the mental constructs of numerical structures, Wertheimer (1912a) prefers to speak of Gebilde.

Is Gestalt psychology representationalist?

More subtle are the arguments put forth by Scheerer. Referring to Khler (1913), he admits (1994) that regarding the phenomenology of perception Khler, like all other Gestaltists, was not representationalist. Qua experience, our perceptual acquaintance with things, persons, and events is not mediated by some "intermediate" entities such as sensations. But if we take a more liberal reading of 'representational theory', reducing it to the mere assumption that our knowledge of the outside world is not direct but mediated by mental entities "standing for" the "physical" entities that form the ultimate objects of knowledge, then Khler's theory is representational

Is Gestalt psychology representationalist?

In other words, according to Scheerer the very fact that we believe in a transphenomenal word is sufficient to argue that our cognitive system is representational. But this means simply that all we can call secondary processes in Kanizsa's sense can be considered representational. But this, according to Khler (1960), are exactly the processes that have no similarities in the phenomenal world. In my opinion, all this indicates clearly that the very concept of representation is rejected in Gestalttheorie. At most we can say that the Gestalt authors had an indifferentist stance on this problem (Luccio, 2003).

Is Gestalt psychology representationalist?

As von Fieandt put it, for Khler psychologically speaking the immediate experiences are in fact what is real. For him, the contents of the directly accessible world dont stay for something else, as the concept of representation would imply, but stay for the contents themselves. This has induced some author (for instance, Keiler, 1980) to accuse the Gestalt psychologists (in the case, Khler) to be a biased idealist . Here it is important to stress the difference that Khler posits between subjective and objective experiences, both results of organic processes (Khler, 1947).

Is Gestalt psychology representationalist?

The subjective experiences are the contents of the phenomenal world that are felt as belonging personally to the subject, and are in so far subjective []. Under no circumstances will it be something merely perceived, or in any sense a subjective phenomenon. (ibidem) Still more clear is Wolfgang Metzger (1941, 2. Kap.), in his classic treatment of the psychic reality [seelisch Wirklichen]. According to Metzger, the first distinction that one must operate is between the physical or metaempirical world [physikalische oder erlebnisjenseitig Welt] and the phenomenal or lived world [anschauliche oder erlebte Welt].

Structural v. functional isomorphism

Isomorphism is structural identity. But here again somebody maintains that it is identity of function for example, Lehar (2003); or Hubbard (2008). King and Michael Wertheimer (2005) too, in their book on Max Wertheimer, say (p. 376, italics added) that Gestalt psychology assumes a kind of functional and structural comparability of phenomenal experiences and of their underlying physiological events. Remember, the similarity must be sachlicher, what function for definition is not.

Structural v. functional isomorphism

In Khler the fact that the isomorphism is structural and not functional is very clear, and it is stated in many instances, directly and indirectly. A functional isomorphism is explicitly denied by Khler in several occasions. In many cases we can observe a correspondence between functional and structural relations (for instance, geometrical relations). But the isomorphism apply to the structure, not to the function.

Structural v. functional isomorphism

The distinction between structure and function will result absolutely clear from an analysis of the additional assumptions that Khler (1938, p. 223 f.) states. In the visual realm, the first assumptions refers to continuity: this is a structural property, that can serve to several phenomenal functions: for instance, as in the case of Wertheimers principle of good continuation (1923), the function is to ensuring the belonginess of a series of points to the same curve.

Structural v. functional isomorphism

We then assume that this continuity corresponds of a continuity of electro-physiological processes in the visual cortex of the person that is experiencing the phenomenal continuity. Of course, this continuity in the physiological processes is structurally identical to the phenomenal continuity, but it is clear that the function that it serves is quite different: for instance, to ensure an apt physico-chemical condition for the development of further physiological processes.

Structural v. functional isomorphism

Also in this case, the Gestalt authors are in part responsible for misunderstandings. So, for instance, Khler (1958, p. 151), writes: Consequently, a given kind of process may occur in different locations; and, if this is true, we have to deal with a functional problem which studies of localization alone cannot solve. My own work in physiological psychology refers to functional questions in this sense. Similar ambiguity one can find in S. E. Asch (1958).

Structural v. functional isomorphism

However, once again Scheerer (1994) invite us to assume a more liberal attitude, and this on two grounds: i) the states of the system cannot be individuated in an absolute fashion but only in relation to other states, ii) evolution works under the physics of dynamical systems and for that reason often has reached one and the same functional solution (e.g. for sensory projection areas) with diverse materials and topographical arrangements.

Monism v. dualism
According to Bunge the Fechnerian parallelism was a dualistic position. The same misunderstanding is also in Bunge for Gestalt position: isomorphism is considered a variant of the parallelism, and so a dualistic position. As a matter of fact, Bunge is not the lone that considered the doctrine of isomorphism an instance of metaphysical dualism. De Laguna (1933) and Boring (1936; see also Boring, 1933) have the same stance . One can add that in the thirties (see King and Max Wertheimer, 2005) there was an intense and respectful correspondence between Boring and Max Wertheimer, the latter trying to convince the first without succeeding of his errors of interpretation.

Monism v. dualism
As Feigl (1934) has clearly stated, isomorphism is an instance of the identity theory, and according to him (Feigl, 1975) Bertrand Russell (with his neutral monism) and Moritz Schlick had the same idea about it. According to Feigl, the problem is to find a language for experience (the language of data) and a language for neural (physiological) events (the physical language), such that we can built sentences in the two languages related to the same events that are isomorphic.

Monism v. dualism
According to Feigl, this means that one can translate every proposition given in one language in a proposition in the other. But mutual translatability means nothing but identity of structure. Logically mutual translatability, isomorphism, means simply identity of the two proposition. Khler(1960, p. 21-22) appears prone to accept Feigls proposal, with some qualification in terms of emergence.

Monism v. dualism
Anyway, Gestalt psychologists were definitely not dualists. As Khler writes, we do not assume, as the philosophers of the past did, that the mind and the body are two substances (1960, p. 3); and we are less and less inclined to believe that the dualistic view can be accepted as final (1960, p. 4). This necessarily means that they endorsed a monistic view, as Newman (1969, 1989) says about Wertheimers stance?

Monism v. dualism
Newman in this is very clear: My main thesis is that the essence of Gestalt psychology [] lies in its insistence upon a monistic approach to psychological fact upon a kind of physicalism, if you will, and a rejection once and for all of dualistic premises, however deceptive may be the packaging (1969). My hypothesis is that the things are more complicated. As note Stadler and Kruse (1994), in 1920 Khlers language was often parallelist. And Metzger (1941, p. 286 f.) endorse a clear parallelist position.

Back to Spinoza
What are in reality the relationships between Gestalt isomorphism and Spinozas identity theory? We know (see Luchins and Luchins, 1982; King and Michael Wertheimer, 2005) that, via his grandfather Jakob Zwicker, yet child Wertheimer, about 10 years old, was acquainted with Spinozas ideas. Wertheimers reading of Spinoza probably offered to him a glimpse of the power of holism (King and Michael Wertheimer, 2005, p. 42), holism that had to be a constant landmark of all Gestalt psychology. We know that all the Weltanschauung of Wertheimer was imprinted by this precocious Spinozian influence.

Back to Spinoza
Furthermore, there is a direct line from Maimonides [Moshe ben Maimon, Rambam], the great Jew theologician of the XII century, and Spinoza. Max Wertheimer and his brother Walter were strongly encouraged by the parents to read religious books during their education. It is quite possibile therefore that Wertheimer knew Maimonides, which was a milestone of the religious education in Jewish families at the time.

Conclusion
1. The concept of isomorphism constitutes an epistemological break with reference to the preceding forms of psychophysical parallelism; The name is unfortunate: this is not a case of the isomorphism in a technical sense; We must restrict the isomorphic domains to phenomenal world and brain processes, beyond the proximal stimulation level; The contents of the phenomenal world cannot be considered in no way representations; The isomorphism is between structures, not functions;

2. 3.

4. 5.

Conclusion
6.
The doctrine of the isomorphism can be considered a sort of neutral monism, neither a dualism nor a monism, materialistic or idealistic, in a Cartesian sense; it is close to the identity in Feigls sense; Many points in this doctrine can be traced back to Spinoza, if not Maimonides; however, it is difficult to argue that a direct conscious reference to Spinoza was present in Gestalt authors; Last but not least, many ambiguities that originated errors of interpretation were caused by ambiguities that are as a matter of fact present in the gestalt writings.

7.

8.